hw4

library(magrittr); library(dplyr); library(MRIcloudT1volumetrics);library('tidyr');library(ggplot2);library(plotly)

Q2

dat = read.table('https://raw.githubusercontent.com/bcaffo/ds4bme/master/data/classInterests.txt',
                 header = 1)
p1 = plot_ly(dat, x = ~Year) %>% add_histogram(name = ~'Year')
p2 = plot_ly(dat, x = ~Program)  %>% add_histogram(name = ~'Program')
subplot(p1, p2)

Q3

library(ggmosaic)
dat = read.table('https://raw.githubusercontent.com/bcaffo/ds4bme/master/data/classInterests.txt'
                 ,header = 1)
g = ggplot(data = dat) +
  geom_mosaic(aes(x = product(Year,Program), fill=Year), na.rm=TRUE) +
  labs(x="Program", y = "Year",title='Class Data') +
  theme(axis.text.x = element_text(angle = 60, hjust = 1))
ggplotly(g)

Q5

dat = read.csv('https://raw.githubusercontent.com/jhu-advdatasci/2018/master/data/KFF/healthcare-spending.csv',skip = 2,header = 1)[2:52,]
colnames(dat)[-1] = c(1991:2014)
dat = gather(dat,key = Year, value = Spending, 2:25) 
g = ggplot(dat, aes(x=Year,y=Spending,col=Location,group = Location)) + 
  geom_line()+
  theme(axis.text.x = element_text(angle = 90,vjust=0))
ggplotly(g)

Q6

dat = read.csv('https://raw.githubusercontent.com/jhu-advdatasci/2018/master/data/KFF/healthcare-spending.csv',skip = 2,header = 1)[2:52,]
dat = mutate(dat,Spending = apply(dat[-1],1,mean)) %>% select(Location,Spending)
g = ggplot(dat,aes(x=Location,y = Spending)) + 
  geom_bar(stat = "identity")+
  theme(axis.text.x = element_text(angle = 90, hjust = 1,vjust = 0))+
  labs(x = "States")
ggplotly(g)